Penn State's Atom-Thin CrOCl Achieves 200:1 Plasma-Etch Selectivity, Outperforming Every Conventional Hard Mask

By NineScrolls Team · 2026-03-25 · 3 min read · Industry

The Breakthrough: A 2D Material That Plasma Cannot Etch

Researchers at Penn State University have published findings in Nature Materials showing that chromium oxychloride (CrOCl), a two-dimensional van der Waals crystal, resists fluorine-based plasma etching far better than any hard mask material currently used in semiconductor manufacturing. The discovery, led by Saptarshi Das, the Ackley Professor of Engineering Science, was reported on March 10, 2026.

The finding was accidental. The team originally attempted to etch CrOCl for an unrelated project and found the material essentially refused to be removed. When bombarded with SF6/O2 plasma — the standard chemistry for etching silicon — CrOCl forms a self-limiting passivation layer on its surface that blocks further material removal. This behavior has not been observed in any conventional hard mask material, including amorphous carbon, silicon nitride, silicon dioxide, aluminum oxide, chromium, nickel, or titanium nitride.

By the Numbers: 30x Better Than Silicon Nitride

The performance gap is not incremental. CrOCl achieves etch rates as low as approximately 2.4 nm/min under aggressive SF6/O2 plasma, yielding an etch selectivity greater than 200:1 relative to silicon. That represents roughly a 30x improvement over silicon nitride (Si3N4), a 20x improvement over titanium nitride (TiN), and a 2.3x improvement over aluminum oxide (Al2O3) under identical conditions. A related material, iron oxychloride (FeOCl), showed similar resistance.

Using CrOCl masks, the team demonstrated deep silicon etching with aspect ratios exceeding 39:1 and minimal feature distortion — a critical capability for fabricating the high-aspect-ratio structures found in 3D NAND flash, advanced logic transistors, and through-silicon vias.

The Unexpected Bonus: Plasma Makes It Smoother

In conventional hard masks, repeated plasma exposure roughens the mask surface, causing micro-masking defects that degrade the pattern transferred into the underlying silicon. CrOCl behaves in reverse. Its layered van der Waals structure means that plasma bombardment preferentially removes rough surface regions, exposing smoother layers beneath. The result is sharper, more vertical sidewalls in the etched features.

The material can also be patterned separately and transferred onto delicate substrates such as flexible plastics or glass, opening potential applications in flexible electronics, sensors, and MEMS devices beyond conventional silicon wafer processing.

What This Means for Plasma Processing and Thin Film Equipment

If CrOCl hard masks reach production, the implications for plasma processing equipment are significant. Etch chambers running fluorine-based chemistries — the workhorses of companies like Lam Research, Tokyo Electron, and Applied Materials — would need to be validated against the new mask material. Process recipes, endpoint detection algorithms, and chamber seasoning protocols would all require adjustment.

For thin film deposition systems, the opportunity is equally direct. Producing uniform, wafer-scale CrOCl films will require precision deposition methods. Chemical vapor deposition (CVD), atomic layer deposition (ALD), or physical vapor deposition (PVD/sputtering) of van der Waals oxyhalides is largely unexplored at manufacturing scale, representing a potential new market for deposition equipment makers and precursor suppliers. Plasma-enhanced CVD (PECVD) systems, which already dominate many thin film steps in semiconductor fabs, would be natural candidates for scaling CrOCl growth.

The equipment supply chain — plasma sources, gas delivery systems, vacuum components, and process monitoring instruments — would need to adapt to handle the new precursor chemistries involved in depositing chromium oxyhalide films.

Limitations and the Road to Fab Adoption

The research so far has been demonstrated only on small exfoliated flakes, not on full wafers. Scaling CrOCl to uniform wafer-scale production is the critical remaining challenge before any fab can consider adoption. The team, which includes collaborators from the University of Chemistry and Technology in Prague, was funded by the National Science Foundation, the U.S. Office of Naval Research, and the U.S. Office of Army Research.

Doctoral candidates Ziheng Chen and Pranavram Venkatram were the study's first authors. The work represents an early but potentially disruptive step in the ongoing effort to push plasma etch processes deeper into the angstrom era.